Anomaly inspection in sewer pipes using stereo vision

Abstract

This paper describes and evaluates a new sensing methodology for anomaly inspection in sewer pipes using a machine vision approach. Currently, most commercial pipe inspection systems include a mobile remote platform and one or more Closed Circuit Television (CCTV) cameras. These inspection systems are slow, costly and have limited accuracy (caused by humans or environmental reasons). More sophisticated approaches (Laser-based, Infra-red Thermography, Ultrasonic-based and Ground Penetrating Radar) suffer from, lack of resolution and an inability to detect water inflow. The main objective of this research is to apply stereo vision technique to generate 3D images of anomalies in sewer pipes in order to achieve high efficiency and accuracy for pipe inspection. The results showed that various types of defects were successfully reconstructed for later advanced processing.